Looking back, looking ahead
February 6, 2008
This post is aimed at serving as something of a self evaluation of the first 6 months of this blog, and a statement of intent for its future.
Two things I would like to do in the near future have to do with the past: First, I have a significant number of drafts for posts that have been stuck in that stage for a while and seem to have little chance of becoming full fledged posts. Those drafts make various points regarding topics I discussed, but apparently I don’t have enough coherent substance in my mind about those matters, so, at this point, I will be satisfied with making those points briefly without substantiation. I would therefore like to clear those posts away by creating a couple of omnibus posts, one in each main area I dealt with, in which the idea of each draft post is briefly mentioned.
The second task is to create a couple of summary or annotated index posts, again, one in each main area I dealt with, detailing the main points made in the posts so far, and linking to the relevant posts.
For both of those tasks, I need to identify the areas of occupation of this blog in its first 6 months. The overarching theme of those areas is the idea of anti-elitism: the notion that no one is significantly more worthy than other people. Under this theme, there are two loosely connected areas:
- The main area of occupation of Pro Bono Statistics so far was a theoretical analysis of democracy. This project is running out of steam: it seems to me I have said much of what I had to say on the matter, and some points I probably made several times over. I do have a few additional points that I wanted to mention, which are laid out in some of the draft posts mentioned above.
- The other substantial area covers the closely related topics of science, academia, peer-review and the economy of attention. To a certain extent this may be seen as an inquiry into social aspects of epistemology.
Having explored theoretically the two topics above, I would like to shift the emphasis to more empirical matters (of course, empirical examinations cannot be carried out without a theoretical framework, and so some theory will always be involved). Ideally, I would like the empirical investigations to relate to the theme of anti-elitism and specifically to the areas of democracy and epistemology. However, I would be happy even with an indirect connection, or even questions that are not connected at all to anti-elitism, but are of intellectual interest and of some practical significance.
The series of posts on the studies of mortality in Iraq is within that category of topics, and I hope to continue to pursue this topic. Beyond the immediate practical interest in finding out the truth about reality in Iraq, this topic has other interesting ramifications (see the points at the end of this post).
Specific additional topics of this kind that I have in mind are:
- Anthropogenic global warming mitigation policy: critical examination of green technology, carbon offsets, nuclear power generation; distributional aspects of mitigation policy,
and several topics related to economics. Quantitative economics seems to me to be enjoying unreasonable credibility – important policy decisions are justified based on (although probably not determined by) economic analysis that is quite dubious, but has the appearance of rigor achieved through its use of quantitative models. I would like to examine this issue in general and various topics in particular:
- Free markets – what exactly are the optimality claims that are made by classical theory about “free markets”? How are “free markets” defined? How do any theoretical models stand up to empirical examination? Can a war induced economic boom (such as was supposedly induced by World War II) be compatible with the assumption of economic efficiency?
- Econometrics – Freakonomics has pushed econometrics into semi-popular status. How rigorous are the standard methods of econometrics? How credible are the sweeping statements made by Levitt and his like? How and why are those statements arrived at?
- Behavioral economics – I suspect that this field is simply a grab-bag of “effects”, many of which (if not all) reflect the viewpoint of the economists rather than defects in the rationality of the subjects. Is there a coherent theory of “behavioral economics”? What are the implications, theoretical and political, of ascribing irrational behavior to the average person?
- The consumer price index – the concept of the CPI is ubiquitous in economic analysis, and is usually taken as a statistic that is objectively measurable with high accuracy. How credible is the concept? How accurately can it be measured? What are the political implications of its definition, measurement and use?
In addition, I would like to make some posts on fundamental issues in statistics. Online discussions (mainly in comments in Deltoid) spurred me in the past to write some posts with definitions and some analysis of confidence intervals. However, more discussion made me realize that questions such as “what is a confidence interval good for?” and “what is a maximum likelihood estimate good for?” are actually of more interest than formal definitions. These questions are also less easily answered and deserve careful attention, since it is the those answers that justify (or fail to justify) the use of the formal devices. By the way, I now find it odd that this kind of questions did not get close scrutiny during my (extensive) formal education in statistics.